- Email: [email protected]
Evaluation of human recombinant bone morphogenetic protein-2-loaded tricalcium phosphate implants in rabbits’ bone defects
Bone Vol. 25, No. 2, Supplement August 1999:55S–58S
Evaluation of Human Recombinant Bone Morphogenetic Protein-2-Loaded Tricalcium Phosphate Implants in Rabbits’ Bone Defects PH. LAFFARGUE,1,2 H. F. HILDEBRAND,1 M. RTAIMATE,2 P. FRAYSSINET,3 J. P. AMOUREUX,4 and X. MARCHANDISE1,5 1 Laboratoire de Biophysique, Unite´ Programme´e de Recherche et d’Enseignement Scientifique, Equipe d’Accueil (UPRES EA) 1049, Faculte´ de Me´decine, Lille, France 2 Clinique d’Orthope´die et de Traumatologie, Hoˆpital Roger Salengro, Centre Hospitalier Re´gional et Universitaire (CHRU) de Lille, Lille, France 3 Laboratoire Bioland, Toulouse, France 4 Laboratoire de Dynamique et Structure des Mate´riels Mole´culaires, Universite´ des Sciences et Technologies de Lille (USTL), Universite´ de Lille-I, Villeneuve d’Ascq, France 5 Service Central de Me´decine Nucle´aire et d’Imagerie Fonctionnelle, Hoˆpital Roger Salengro, CHRU de Lille, Lille, France
Key Words: Bone substitute; Osteoinduction; Bone morphogenetic protein; Tricalcium phosphate; Dual-energy X-ray absorptiometry; 31-phosphorus NMR spectroscopy.
Porous -tricalcium phosphate (TCP) has osteoconductive properties. The adsorption of human recombinant bone morphogenetic protein-2 (rhBMP-2) onto TCP could realize an osteoinductive bone substitute. We evaluated it on an animal model using dual-energy X-ray absorptiometry (DEXA) and solid-state 31P nuclear magnetic resonance (NMR) spectroscopy. TCP cylinders loaded with rhBMP-2 were implanted into rabbits’ femoral condyle bone defects, and TCP alone as control into the controlateral femur. We studied two different doses of rhBMP-2 (10 and 40 g) on two groups of four animals. Evaluation consisted in radiography, histology, and histomorphometry, DEXA, and NMR spectroscopy using an original method of quantification. With both doses of rhBMP-2, we observed on radiographs an increase of trabecular bone around implants. Histology showed resorption of the ceramic, trabecular bone with osteoblasts and osteoid substance around the implants, and colonization inside the porous TCP by new bone formed. Histomorphometry showed that the osteoid surface (OS/BS) was greatest with the high dose of rhBMP-2. The difference was slight between the low dose of rhBMP-2 and control. DEXA showed a dose-dependent increase of bone mineral density of rhBMP2-loaded TCP vs. control. NMR spectroscopy confirmed that the amount of new bone formed in TCP was greater when TCP carried rhBMP-2, and increased with the dose of rhBMP-2 used. We showed that TCP was a good matrix for rhBMP-2, which gave it osteoinductive properties in an orthotopic site, in a dose-dependent manner. Thus, such composite biomaterial seems to be of great interest in reconstructive bone surgery. Further studies are needed in clinical practice to determine optimal doses. (Bone 25:55S–58S; 1999) © 1999 by Elsevier Science Inc. All rights reserved.
Introduction New composite osteoinductive biomaterials have great potential for reconstructive surgery of bone defects. Recombinant human bone morphogenetic protein-2 (rhBMP-2) carried onto demineralized bone matrix, collagen, or poly(␣-hydroxyacids) is known to have osteoinductive properties.5,7,8 One limiting factor to its use is the availability of delivery matrices. -tricalcium phosphate (TCP) is a biodegradable synthetic ceramic, with great bioactivity and biocompatibility.1,9 We used TCP as a nonimmunogenic, inorganic carrier for rhBMP-2 to realize a composite biomaterial and we studied osteoinduction in rabbits’ femoral bone defects. Usual evaluation methods (standardized radiography, histology, and histomorphometry) were completed by two quantitative methods: dual-energy x-ray absorptiometry (DEXA)4 and solid-state 31-phosphorus nuclear magnetic resonance (NMR) spectroscopy.2,3 Material and Methods Implants TCP (Bioland, Toulouse, France), 100% pure, 70% interconnected macroporosity (pores of 200- to 500-m diameter) was used as cylinders 5 mm in diameter and 10 mm long. Lyophilized rhBMP-2 (Genetics Institute, Cambridge, MA) was reconstituted in sterile water then diluted in MFR-842 buffer to obtain two solutions of different concentration, according to Genetics Institute dilution schema. Each rhBMP-2 solution was dripped onto the TCP cylinders, taking care to wet them uniformly, to obtain two composite biomaterials with either 10 or 40 g of rhBMP-2. Operative Procedure
Address for correspondence and reprints: Dr. Philippe Laffargue, Clinique d’Orthope´die et de Traumatologie, Hoˆpital Roger Salengro, CHRU de Lille, 59037 Lille Cedex, France. © 1999 by Elsevier Science Inc. All rights reserved.
Eight mature male New Zealand White rabbits were cared for following the European (OEDC) requirements for “Good Labo55S
8756-3282/99/$20.00 PII S8756-3282(99)00134-9
56S
Ph. Laffargue et al. rhBMP-2-loaded TCP implants evaluation
Figure 1. Experimental and simulated spectra of removed implants, in Group 1, for TCP alone (a), and for rhBMP-2-loaded TCP (10 g) (b). Arrow indicates the bone spectrum.
Bone Vol. 25, No. 2, Supplement August 1999:55S–58S
Figure 2. Experimental and simulated spectra of removed implants, in Group 2, for TCP alone (a), and for rhBMP-2-loaded TCP (40 g) (b). Arrow indicates the bone spectrum.
Bone Vol. 25, No. 2, Supplement August 1999:55S–58S
ratory Practice.” After general anaesthesia, a lateral approach of distal femur was performed aseptically and a cylindrical bone defect (5 mm in diameter, 10 mm long) was carefully manually drilled in femoral condyles. For each rabbit, implants was gently inserted in the defect: rhBMP-2-loaded TCP using the two doses, 10 g (Group 1, n ⫽ 4) and 40 g (Group 2, n ⫽ 4), in the right femur, and TCP alone as control in the left femur. Animals were killed on day 30 after operation.
Ph. Laffargue et al. rhBMP-2-loaded TCP implants evaluation
57S
Histology and Histomorphometry Bone trabeculae were present at the surface of all implants. They were surrounded by osteoid substance bordered by osteoblasts. TCP resorption was seen with ceramic particles in histiocytes and giant cells. In Group 1, no difference was seen for the criteria studied between TCP alone and rhBMP-2-loaded TCP. In Group 2, osteoid surface was higher in rhBMP-2-loaded TCP (OS/BS ⫽ 62%) than in TCP alone (OS/BS ⫽ 11.3%). No significant difference was seen for other criteria.
Evaluation Anteroposterior and lateral radiographs were performed postoperatively under anesthesia and repeated after death, after removing patella and soft tissues. Interpretation was made by three trained independent observers, with only one knowing methodology used. Histological and histomorphometrical studies were performed for one rabbit of each group, with determination of bone surface (BS/TV), osteoid surface (OS/BS), osteoblast surface (Ob.S/BS), and eroded surface (ES/BS).6 DEXA was performed under anesthesia, using a Sophos densitometer, at day 1 and day 30. For determination of the bone mineral density (BMD, g/cm2), a region of interest (ROI) centered on the implant was manually drawn, and another ROI of same size was chosen outside the implant for background determination. A well-known density control was tested at the same time. A third measurement of BMD was done on the removed implant after death and careful extraction from the condyles, for the three rabbits of each group that were not intended for histology. Solid-state 31P NMR spectroscopy was done using a DXM600 spectrometer (Bruker, Wissembourg) with a 14-Tesla magnetic field and using the magic angle spinning method. Experimental and simulated spectra, obtained by Peakfit software (Jandel-SPSS, Erkrath, Germany), were decomposed in gauss curves: one for bone, unique, centered at 875.295 Hz, with a good coefficient of correlation (r2 ⫽ 0.999), three gauss curves for TCP (r2 ⫽ 0.99), and four (one of which at 875.295 Hz) for combinations of bone and TCP. After meticulous extraction, implants were dried and powdered to be examined by NMR spectroscopy. Removed implants composition was quantitated by fitting spectra to a linear combination of bone and TCP. The area of the gauss curve of bone, expressed as a percent of the total area of the spectrum, gave the percentage of new formed bone in the sample.
Results Complications One fracture of the right femoral shaft (proximal third) occurred 24 h after implantation for one rabbit of Group 2. Bone healing was observed spontaneously without body weight loss during the whole study. No sepsis occurred, nor any else complication. Radiography All implants, with rhBMP-2 or not, were more dense at day 30 than at day 1. In Group 1, the rhBMP-2-loaded implants were most dense in three out of four rabbits. No difference was seen in the fourth one. In Group 2, in three out of four rabbits, the rhBMP-2-loaded implant was more dense and was surrounded by a high number of dense trabeculae. For the fourth rabbit of Group 2 (with postoperative femoral shaft fracture), no difference was seen.
DEXA Mean coefficient of variation was 10% for in vivo measurements and 2% for ex vivo ones. Therefore, differences were significant if higher than 17% in vivo and 3.6% ex vivo. With in vivo measurements, no difference was seen between TCP alone and rhBMP-2loaded TCP, at day 1 and day 30, in the two groups. Ex vivo measurements showed significant BMD increase for the rhBMP-2loaded TCP, compared with control, most significantly in Group 2 than in Group 1 (27.4% and 13%, respectively). RMN Spectroscopy Removed implants gave spectra that combined the bone spectrum and the more complex spectrum of TCP in variable proportions. Experimental and simulated spectra are shown for controls (Figures 1A and 2A) and for rhBMP-2-loaded TCP (10 g in Figure 1B and 40 g in Figure 2B). Proportion of new bone formed was higher in rhBMP-2-loaded implants than in TCP alone, and the difference was higher in Group 2 (18.2% ⫾ 5.3% with 40 g of rhBMP-2 vs. 7% ⫾ 2.9% in control) than in Group 1 (4.3% ⫾ 1.9 with 10 g vs. 2.3% ⫾ 1.5% in control). Discussion Concerning our methodology, only one rabbit of each group was studied by histology and histomorphometry, but no more could be studied because implants had to be pulverized for RMN spectroscopy. Concerning evaluation, radiography is of interest in analyzing the implants and bone reaction around them, but it does not give precisely quantitative data. In our study, DEXA was interpretable only ex vivo because the densitometer did not allow precise measurements in vivo. Our method of quantification by NMR spectroscopy was of great interest to determine chemical transformations, new bone formed inside porous implants, TCP resorption, and TCP-bone ratio variations in the initial defect. Our results confirmed the interest of TCP as a growth factor carrier into bone and the dose-dependent osteoinductive properties of rhBMP-2 in an orthotopic site, as well as osteogenesis around the implants and densification of bone trabeculae. Further studies with evaluations at different time intervals are needed to determine optimal doses of rhBMP-2 and interest of association of bone growth factors, such as insulin-like growth factor-I and rhBMP-2 loaded onto TCP to induce synergetic effects for osteoinduction. Acknowledgments: The authors thank Genetics Institute for kindly providing rhBMP-2 and Bioland for kindly providing -tricalcium phosphate cylinders used in the study.
References 1. Frayssinet, P., Trouillet, J. L., Rouquet, N., Azimus, E., and Autefage, A. Osseointegration of macroporous calcium phosphate ceramics having a different chemical composition. Biomaterials 14:423– 429; 1993.
58S
Ph. Laffargue et al. rhBMP-2-loaded TCP implants evaluation
2. Marchandise, X., Belgrand, P., Laffargue, P., Miquel, J. L., Lemaıˆtre, J., and Legrand, A. P. Spectroscopie RMN des biomate´riaux. Innov Tech Biol Med 16:49 –59; 1995. 3. Marchandise, X., Belgrand, P., and Legrand, A. P. Solid-state 31P NMR spectroscopy of bone and bone substitutes. Magn Reson Med 28:1– 8; 1992. 4. Mosheiff, R., Klein, B. Y., Leichter, I., et al. Use of dual-energy X-ray absorptiometry (DEXA) to follow mineral content changes in small ceramic implants in rats. Biomaterials 13:462– 466; 1992. 5. Nimni, M. E. Polypeptide growth factors: targeted delivery systems. Review. Biomaterials 18:1201–1225; 1997. 6. Parfitt, A. M., Drezner, M. K., Glorieux, F. H., et al. Bone histomorphometry:
Bone Vol. 25, No. 2, Supplement August 1999:55S–58S standardization of nomenclature, symbols, and units. J Bone Miner Res 2:595– 610; 1987. 7. Riley, E. H., Lane, J. M., Urist, M. R., Lyons, K. M., and Lieberman, J. R.Bone morphogenetic protein-2. Biology and applications. Clin Orthop 324:39 – 46; 1996. 8. Rodgers, J. B., Vasconez, H. C., Wells, M. D., et al. Two lyophilized polymer matrix recombinant human bone morphogenetic protein-2 carriers in rabbit calvarial defects. J Craniofac Surg 9:147–153; 1998. 9. Yamada, S., Heymann, D., Bouler, J. M., and Daculsi, G. Osteoclastic resorption of calcium phosphate ceramics with different hydroxyapatite/-tricalcium phosphate ratios. Biomaterials 18:1037–1041; 1997.
Evaluation of Human Recombinant Bone Morphogenetic Protein-2-Loaded Tricalcium Phosphate Implants in Rabbits’ Bone Defects PH. LAFFARGUE,1,2 H. F. HILDEBRAND,1 M. RTAIMATE,2 P. FRAYSSINET,3 J. P. AMOUREUX,4 and X. MARCHANDISE1,5 1 Laboratoire de Biophysique, Unite´ Programme´e de Recherche et d’Enseignement Scientifique, Equipe d’Accueil (UPRES EA) 1049, Faculte´ de Me´decine, Lille, France 2 Clinique d’Orthope´die et de Traumatologie, Hoˆpital Roger Salengro, Centre Hospitalier Re´gional et Universitaire (CHRU) de Lille, Lille, France 3 Laboratoire Bioland, Toulouse, France 4 Laboratoire de Dynamique et Structure des Mate´riels Mole´culaires, Universite´ des Sciences et Technologies de Lille (USTL), Universite´ de Lille-I, Villeneuve d’Ascq, France 5 Service Central de Me´decine Nucle´aire et d’Imagerie Fonctionnelle, Hoˆpital Roger Salengro, CHRU de Lille, Lille, France
Key Words: Bone substitute; Osteoinduction; Bone morphogenetic protein; Tricalcium phosphate; Dual-energy X-ray absorptiometry; 31-phosphorus NMR spectroscopy.
Porous -tricalcium phosphate (TCP) has osteoconductive properties. The adsorption of human recombinant bone morphogenetic protein-2 (rhBMP-2) onto TCP could realize an osteoinductive bone substitute. We evaluated it on an animal model using dual-energy X-ray absorptiometry (DEXA) and solid-state 31P nuclear magnetic resonance (NMR) spectroscopy. TCP cylinders loaded with rhBMP-2 were implanted into rabbits’ femoral condyle bone defects, and TCP alone as control into the controlateral femur. We studied two different doses of rhBMP-2 (10 and 40 g) on two groups of four animals. Evaluation consisted in radiography, histology, and histomorphometry, DEXA, and NMR spectroscopy using an original method of quantification. With both doses of rhBMP-2, we observed on radiographs an increase of trabecular bone around implants. Histology showed resorption of the ceramic, trabecular bone with osteoblasts and osteoid substance around the implants, and colonization inside the porous TCP by new bone formed. Histomorphometry showed that the osteoid surface (OS/BS) was greatest with the high dose of rhBMP-2. The difference was slight between the low dose of rhBMP-2 and control. DEXA showed a dose-dependent increase of bone mineral density of rhBMP2-loaded TCP vs. control. NMR spectroscopy confirmed that the amount of new bone formed in TCP was greater when TCP carried rhBMP-2, and increased with the dose of rhBMP-2 used. We showed that TCP was a good matrix for rhBMP-2, which gave it osteoinductive properties in an orthotopic site, in a dose-dependent manner. Thus, such composite biomaterial seems to be of great interest in reconstructive bone surgery. Further studies are needed in clinical practice to determine optimal doses. (Bone 25:55S–58S; 1999) © 1999 by Elsevier Science Inc. All rights reserved.
Introduction New composite osteoinductive biomaterials have great potential for reconstructive surgery of bone defects. Recombinant human bone morphogenetic protein-2 (rhBMP-2) carried onto demineralized bone matrix, collagen, or poly(␣-hydroxyacids) is known to have osteoinductive properties.5,7,8 One limiting factor to its use is the availability of delivery matrices. -tricalcium phosphate (TCP) is a biodegradable synthetic ceramic, with great bioactivity and biocompatibility.1,9 We used TCP as a nonimmunogenic, inorganic carrier for rhBMP-2 to realize a composite biomaterial and we studied osteoinduction in rabbits’ femoral bone defects. Usual evaluation methods (standardized radiography, histology, and histomorphometry) were completed by two quantitative methods: dual-energy x-ray absorptiometry (DEXA)4 and solid-state 31-phosphorus nuclear magnetic resonance (NMR) spectroscopy.2,3 Material and Methods Implants TCP (Bioland, Toulouse, France), 100% pure, 70% interconnected macroporosity (pores of 200- to 500-m diameter) was used as cylinders 5 mm in diameter and 10 mm long. Lyophilized rhBMP-2 (Genetics Institute, Cambridge, MA) was reconstituted in sterile water then diluted in MFR-842 buffer to obtain two solutions of different concentration, according to Genetics Institute dilution schema. Each rhBMP-2 solution was dripped onto the TCP cylinders, taking care to wet them uniformly, to obtain two composite biomaterials with either 10 or 40 g of rhBMP-2. Operative Procedure
Address for correspondence and reprints: Dr. Philippe Laffargue, Clinique d’Orthope´die et de Traumatologie, Hoˆpital Roger Salengro, CHRU de Lille, 59037 Lille Cedex, France. © 1999 by Elsevier Science Inc. All rights reserved.
Eight mature male New Zealand White rabbits were cared for following the European (OEDC) requirements for “Good Labo55S
8756-3282/99/$20.00 PII S8756-3282(99)00134-9
56S
Ph. Laffargue et al. rhBMP-2-loaded TCP implants evaluation
Figure 1. Experimental and simulated spectra of removed implants, in Group 1, for TCP alone (a), and for rhBMP-2-loaded TCP (10 g) (b). Arrow indicates the bone spectrum.
Bone Vol. 25, No. 2, Supplement August 1999:55S–58S
Figure 2. Experimental and simulated spectra of removed implants, in Group 2, for TCP alone (a), and for rhBMP-2-loaded TCP (40 g) (b). Arrow indicates the bone spectrum.
Bone Vol. 25, No. 2, Supplement August 1999:55S–58S
ratory Practice.” After general anaesthesia, a lateral approach of distal femur was performed aseptically and a cylindrical bone defect (5 mm in diameter, 10 mm long) was carefully manually drilled in femoral condyles. For each rabbit, implants was gently inserted in the defect: rhBMP-2-loaded TCP using the two doses, 10 g (Group 1, n ⫽ 4) and 40 g (Group 2, n ⫽ 4), in the right femur, and TCP alone as control in the left femur. Animals were killed on day 30 after operation.
Ph. Laffargue et al. rhBMP-2-loaded TCP implants evaluation
57S
Histology and Histomorphometry Bone trabeculae were present at the surface of all implants. They were surrounded by osteoid substance bordered by osteoblasts. TCP resorption was seen with ceramic particles in histiocytes and giant cells. In Group 1, no difference was seen for the criteria studied between TCP alone and rhBMP-2-loaded TCP. In Group 2, osteoid surface was higher in rhBMP-2-loaded TCP (OS/BS ⫽ 62%) than in TCP alone (OS/BS ⫽ 11.3%). No significant difference was seen for other criteria.
Evaluation Anteroposterior and lateral radiographs were performed postoperatively under anesthesia and repeated after death, after removing patella and soft tissues. Interpretation was made by three trained independent observers, with only one knowing methodology used. Histological and histomorphometrical studies were performed for one rabbit of each group, with determination of bone surface (BS/TV), osteoid surface (OS/BS), osteoblast surface (Ob.S/BS), and eroded surface (ES/BS).6 DEXA was performed under anesthesia, using a Sophos densitometer, at day 1 and day 30. For determination of the bone mineral density (BMD, g/cm2), a region of interest (ROI) centered on the implant was manually drawn, and another ROI of same size was chosen outside the implant for background determination. A well-known density control was tested at the same time. A third measurement of BMD was done on the removed implant after death and careful extraction from the condyles, for the three rabbits of each group that were not intended for histology. Solid-state 31P NMR spectroscopy was done using a DXM600 spectrometer (Bruker, Wissembourg) with a 14-Tesla magnetic field and using the magic angle spinning method. Experimental and simulated spectra, obtained by Peakfit software (Jandel-SPSS, Erkrath, Germany), were decomposed in gauss curves: one for bone, unique, centered at 875.295 Hz, with a good coefficient of correlation (r2 ⫽ 0.999), three gauss curves for TCP (r2 ⫽ 0.99), and four (one of which at 875.295 Hz) for combinations of bone and TCP. After meticulous extraction, implants were dried and powdered to be examined by NMR spectroscopy. Removed implants composition was quantitated by fitting spectra to a linear combination of bone and TCP. The area of the gauss curve of bone, expressed as a percent of the total area of the spectrum, gave the percentage of new formed bone in the sample.
Results Complications One fracture of the right femoral shaft (proximal third) occurred 24 h after implantation for one rabbit of Group 2. Bone healing was observed spontaneously without body weight loss during the whole study. No sepsis occurred, nor any else complication. Radiography All implants, with rhBMP-2 or not, were more dense at day 30 than at day 1. In Group 1, the rhBMP-2-loaded implants were most dense in three out of four rabbits. No difference was seen in the fourth one. In Group 2, in three out of four rabbits, the rhBMP-2-loaded implant was more dense and was surrounded by a high number of dense trabeculae. For the fourth rabbit of Group 2 (with postoperative femoral shaft fracture), no difference was seen.
DEXA Mean coefficient of variation was 10% for in vivo measurements and 2% for ex vivo ones. Therefore, differences were significant if higher than 17% in vivo and 3.6% ex vivo. With in vivo measurements, no difference was seen between TCP alone and rhBMP-2loaded TCP, at day 1 and day 30, in the two groups. Ex vivo measurements showed significant BMD increase for the rhBMP-2loaded TCP, compared with control, most significantly in Group 2 than in Group 1 (27.4% and 13%, respectively). RMN Spectroscopy Removed implants gave spectra that combined the bone spectrum and the more complex spectrum of TCP in variable proportions. Experimental and simulated spectra are shown for controls (Figures 1A and 2A) and for rhBMP-2-loaded TCP (10 g in Figure 1B and 40 g in Figure 2B). Proportion of new bone formed was higher in rhBMP-2-loaded implants than in TCP alone, and the difference was higher in Group 2 (18.2% ⫾ 5.3% with 40 g of rhBMP-2 vs. 7% ⫾ 2.9% in control) than in Group 1 (4.3% ⫾ 1.9 with 10 g vs. 2.3% ⫾ 1.5% in control). Discussion Concerning our methodology, only one rabbit of each group was studied by histology and histomorphometry, but no more could be studied because implants had to be pulverized for RMN spectroscopy. Concerning evaluation, radiography is of interest in analyzing the implants and bone reaction around them, but it does not give precisely quantitative data. In our study, DEXA was interpretable only ex vivo because the densitometer did not allow precise measurements in vivo. Our method of quantification by NMR spectroscopy was of great interest to determine chemical transformations, new bone formed inside porous implants, TCP resorption, and TCP-bone ratio variations in the initial defect. Our results confirmed the interest of TCP as a growth factor carrier into bone and the dose-dependent osteoinductive properties of rhBMP-2 in an orthotopic site, as well as osteogenesis around the implants and densification of bone trabeculae. Further studies with evaluations at different time intervals are needed to determine optimal doses of rhBMP-2 and interest of association of bone growth factors, such as insulin-like growth factor-I and rhBMP-2 loaded onto TCP to induce synergetic effects for osteoinduction. Acknowledgments: The authors thank Genetics Institute for kindly providing rhBMP-2 and Bioland for kindly providing -tricalcium phosphate cylinders used in the study.
References 1. Frayssinet, P., Trouillet, J. L., Rouquet, N., Azimus, E., and Autefage, A. Osseointegration of macroporous calcium phosphate ceramics having a different chemical composition. Biomaterials 14:423– 429; 1993.
58S
Ph. Laffargue et al. rhBMP-2-loaded TCP implants evaluation
2. Marchandise, X., Belgrand, P., Laffargue, P., Miquel, J. L., Lemaıˆtre, J., and Legrand, A. P. Spectroscopie RMN des biomate´riaux. Innov Tech Biol Med 16:49 –59; 1995. 3. Marchandise, X., Belgrand, P., and Legrand, A. P. Solid-state 31P NMR spectroscopy of bone and bone substitutes. Magn Reson Med 28:1– 8; 1992. 4. Mosheiff, R., Klein, B. Y., Leichter, I., et al. Use of dual-energy X-ray absorptiometry (DEXA) to follow mineral content changes in small ceramic implants in rats. Biomaterials 13:462– 466; 1992. 5. Nimni, M. E. Polypeptide growth factors: targeted delivery systems. Review. Biomaterials 18:1201–1225; 1997. 6. Parfitt, A. M., Drezner, M. K., Glorieux, F. H., et al. Bone histomorphometry:
Bone Vol. 25, No. 2, Supplement August 1999:55S–58S standardization of nomenclature, symbols, and units. J Bone Miner Res 2:595– 610; 1987. 7. Riley, E. H., Lane, J. M., Urist, M. R., Lyons, K. M., and Lieberman, J. R.Bone morphogenetic protein-2. Biology and applications. Clin Orthop 324:39 – 46; 1996. 8. Rodgers, J. B., Vasconez, H. C., Wells, M. D., et al. Two lyophilized polymer matrix recombinant human bone morphogenetic protein-2 carriers in rabbit calvarial defects. J Craniofac Surg 9:147–153; 1998. 9. Yamada, S., Heymann, D., Bouler, J. M., and Daculsi, G. Osteoclastic resorption of calcium phosphate ceramics with different hydroxyapatite/-tricalcium phosphate ratios. Biomaterials 18:1037–1041; 1997.